isprsarchives XXXVIII 6 W27 85 2011
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
GIS AWARENESS FOR SURVEYORS –
A SPATIAL DATA DISTRIBUTION PERSECTIVE
Jung-Hong Hong a, *, Hone-Jay Chu a
a
Dept. of Geomatics, National Cheng-Kung University, Tainan, Taiwan – (junghong, honejay)@mail.ncku.edu.tw
KEY WORDS: surveyor, education, digitizing, geospatial data, metadata
ABSTRACT:
The rapid technological growth of GIS, internet and mobile devices has steered a variety of innovated and pioneering applications in
the past few years. All of these applications require spatial data that meet specified criteria to ensure the quality of the service content.
The traditional role for surveyors was to use surveying technology and instruments to determine the location of phenomena of
interests and offer map products to serve as the basis for decision making. As the application demands becomes more variety and
instantaneous, the role and responsibility for modern surveyors must be examined. Professional surveyors nowadays must have both
the knowledge and skills to produce geospatial data that precisely meet the various application needs and correctly convey their
understanding about data to domain users to avoid wrong decision making. This paper discussed how we include such considerations
into our curriculum design at the Dept of Geomatics, NCKU. During their studies, students are expected to establish a correct
understanding about the role and responsibility of surveyors and develop themselves to be “solution providers” for providing reliable
geospatial data.
1.
applications because it has dominant influence on the use of the
application. To produce geospatial data that precisely meet the
application needs is always regarded as the responsibility and
the first priority to surveyors. Surveyors are thus required to
receive all kinds of professional training, such that they have
sufficient professional knowledge to make a survey plan and
ensure the criteria about the quality of the products can be met.
A rather restricted viewpoint about the responsibility of
surveyors is surveyors turn in the final products (e.g., maps) and
leave the remaining works to the organizations that intend to
use the data. In the current data sharing environment, this
viewpoint is no longer valid because users nowadays can easily
search “all” of the available geospatial data, regardless of what
their original purpose is. Despite that datasets are still created
according to specific purposes, there is no way to predict how it
is going to be used. It becomes very risky if users cannot have a
correct understanding about the data they access and use. From
the perspective of data production, it is hence necessary to
examine how surveyors cope with the vast varieties and fast
growing of needs from the geospatial community. This paper
discusses our viewpoint regarding the role and responsibility
change of surveyors, and further uses the lab of the digitization
of geospatial data as an example to demonstrate how this
viewpoint is incorporated in our curriculum design. The
remaining of this paper is organized as follows: section 2 offers
a brief overview about the GIS education in the Dept. of
Geoamtics at NCKU and our viewpoints about the links
between surveyors and GIS-based applications. Section 3 uses
the topics of digitizing and distribution of geospatial data in the
GIS lab as an example to demonstrate our design approach.
Finally, section 4 summarized our major findings.
INTRODUCTION
The rapid technological growth of GIS, internet and mobile
devices has steered a variety of innovated and pioneering
applications in the past few years. For example, even an
inexperienced user can now easily access the electronic maps
and satellite images of an unfamiliar place via internet-based
service platforms like Google Earth or Bing Maps. As internet
resources provide “free” information to unrestricted domains of
users in such a convenient way, the education of geomatics,
regardless what levels they are, have forever changed. Another
interesting example is the overwhelming expansion of mobile
telecommunication markets, especially the success of those
technology based upon smart phones. By simply downloading
an APP, we can use smart phones to easily acquire location
information with built–in GPS chip and record other types of
information with other built-in devices like camera, voice
recorder or touchpad. For many applications, this is already
sufficient to serve as a mobile device for collecting geospatial
data, not to mention it can be also used in Location-Based
Service (LBS), where service content from remote servers are
retrieved based on the current location of users via internet
connection. These technology innovations surely give a
tremendous boost to the evolution of education in schools, as
the use of data, software and hardware all become much more
convenient and inexpensive than before. The most important
breakthrough is that this impact is not restricted to professional
people who dedicate to the theory, implementation and
applications of geospatial technology, any people, any
department, any organization who has the need of using
geospatial data can all be benefited by the new technology.
Geospatial data is a common and essential component that
needs to receive tremendous attention in the above innovated
* Corresponding author.
85
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
2. CORE OF CURRICULUM DESIGN
Year
sophomore
sophomore
junior
junior
junior
junior
2.1 Brief history of the Dept. of Geomatics at NCKU
The Dept. of Geomatics at NCKU was formerly the Dept. of
Surveying Engineering. Established in the year of 1978, Dept.
of Surveying Engineering was dedicated to the education of
professional surveyors. Students receive solid training from a
variety of surveying courses, e.g., plane surveying, geodesy,
photogrammetry, cartography, etc., to enrich their professional
knowledge and become professional surveyors. As the scope of
geospatial technology evolves with the progress of internet,
remote sensing and GIS, more advanced courses were added
into the curriculum framework. In 2004, the title of the
department was officially changed to the Dept. of Geomatics to
vividly reflect our major missions for dedicating to the
education and training of geospatial professionals. The current
domain of interests can be generally categorized into three
major domains: (1) geodesy and satellite positioning; (2)
photogrammetry and remote sensing and (3) geographic
information system. Students are required to take fundamental
courses from all three domains, and are encouraged to further
pursue advanced studies in any fields they choose. Every
domain has its own course map for students to follow. With a
strong and excellent foundation in surveying engineering
education, becoming a licensed surveyor is often regarded as a
minimal requirement for the alumni of this department. Students
who have strong commitments and research potential to pursue
advanced knowledge about geomatics can continue their
students in the master and Ph D. programs.
Course title
Introduction to cartography
Basic principles of GIS
Applications of GIS
Digital Mapping
GIS Lab
GIS data quality, standards and sharing
Table 1. GIS-related courses
According to the course maps of the department, students are
expected to complete all the required courses of the three major
domains after their junior years. Some students may continue to
pursue advanced studies in GIS fields under the supervision of
teachers. These students are trained to be able to apply for job,
e.g., project manager, engineer, etc., offered by private GIS
consulting firms. Other students may choose to pursue their
career as surveyors. Based on the training from GIS-related
courses, they are able to use GIS software as tools for the
handling and processing of geospatial datasets. Regardless what
their choice of professional career is, a correct understanding
about the production and distribution of geospatial data is the
key to our curriculum design.
2.3 Vision
Via the designed courses, every student graduated from this
department is expected to have sufficient professional
knowledge and ample hand-on experience to become a
professional licensed surveyor. From the perspective of GIS,
every student must have a clear and unambiguous
understanding about how geospatial data is produced and
distributed in the future data sharing environment. Our goals are
therefore designed as the following:
2.2 GIS curriculum
To suffice the needs of GIS training, various topics of GISrelated courses are offered in undergraduate studies (Table 1).
Undergraduate students are required to take at least 6 credits of
hour of courses from each domain to ensure they have broad
knowledge about how geospatial technology is developed and
evolved. Since the other two domains are mainly focusing on
the acquisition and processing of geospatial data, a link between
courses of these two domains and those of GIS must be
established and emphasized beforehand. The course
“Introduction to Cartography” serves as the foundation for all
applications that use maps as the illustration tools. By
introducing the basic principles of map making, students learn
how to design their own maps that can adapt to application
needs and produce maps following predefined mapping
specifications. The course of “Basic principles of GIS” offers
the basic understanding about GIS. Following the lifecycles of
GIS development, we respectively introduce the knowledge
about the acquisition, editing, processing, management, analysis,
distribution and application of geospatial data. Based upon this
basis, the course of “GIS applications” further expands
students’ views regarding how geospatial data can be integrated
and used in different applications. In addition to these three
required courses, the course of “Digital mapping” is designed to
address the use of computers on the production of maps in
digital format. The course of “GIS Lab” is designed to offer
more practical training on different aspects of GIS. The course
of “GIS data quality, standard and sharing” focuses on the
knowledge to facilitate the sharing of geospatial data via
standardized technology Students are expected to develop
themselves to be solution providers and team leaders that can
actively evaluate the best strategy to the given tasks.
1. Students must have the ability to analyze the
requirement of tasks at hands with professional knowledge,
2. Students must be able to offer the best strategy by
taking advantages of tools available to them, both
surveying and GIS.
3. Students must be able to evaluate the quality of their
products and the control of error factors during data
production.
4. Students must be able to convey their understanding
about data to users to avoid wrong decision making.
As an individual course may cover one or two aspects only, we
believe a more comprehensive design that encompass all the
above four goals in a logical and integrated fashion needs to be
introduced in our curriculum. Section 3 demonstrates how we
incorporate this requirement in our GIS lab design.
3. CURRICULUM DESIGN
Based on the visions raised in section 2.3, the following
discusses the lab design for “Digitizing of geospatial data”.
Digitizing was one of the fundamental GIS skills that allow
users to extract the spatial information of phenomena of
interests from existed medium, e.g., maps and imagery (Chang,
2003). The digitized location information is stored in digital
formats for further processing in GIS-based applications.
Typical examples in the earlier years were mainly paper maps,
e.g., historical maps, topographic maps, cadastral maps, etc.
After remote sensing images become powerful resources for
collecting instantaneous information of reality and reliable
reference for analyzing changes over time, the ability of
digitizing becomes a mandatory skill to GIS users. However,
86
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
largely restrict the outcomes of digitizing. As most of the
imagery is produced following specific technical specification,
commonly used imagery types in Taiwan are introduced.
most users simply learn how to “use” the digitizing function
without a correct understanding about the factors that may
influence the quality of the products. To educate surveyors to
offer high-quality digitized geospatial data for later applications,
the following factors are included in this lab:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Google Earth
Google Earth has received increasing attention from both
geospatial professionals and ordinary users becasue it not only
provides global coverage of the earth, but also historical
archives of images at the same place. Furthermore, the use of
Google Earth is free to the public. Google Earth thus quickly
becomes a popular and inexpensive source from which digitized
information can be collected. Despite its advantages, the
drawback for Google Earth is its inconsistent quality at different
places and users generally are not offered sufficient information
to determine its fitness for use. When users are not sure whether
the referenced material is reliable or not, the digitized result is
also questionable.
The basic workflow of digitizing process
The determination of coordinate reference system
The choose of common control points
The geo-registration procedures
The distinct property of various types of data sources
The efficient digitizing procedures
The evaluation of positional accuracy
The topological constraints of digitized data
The completeness constraints of digitized data
The use of free internet resources
The establishment of standardized metadata
The digitizing process in this lab is composed of four
major steps (Figure 1):
Web Map Service
Web Map Service (OGC, 2006) has quickly become an easy
way for organizations to publish their data to the internet. With
its imagery-based nature, users can easily acquire data from the
servers in imagery format and digitize phenomena of interests
with tools available from clients’ software. As WMS requires
providers to clearly specify its referenced CRS, the procedure is
very similar to the procedure for digitizing a RS image. Unlike
the above three types of sources, the image returned can be
arbitrarily specified by users and can be even superimposed in
Google Earth for further comparison. The National Center for
High-Performance Computing (NCHC) provides a variety of
WMS based on the RS images it collected. Unlike Google Earth,
the WMS from NCHS is dedicated to the images of Taiwan.
Some of them are the images collected after major natural
hazards. This serves as an outstanding resource for students to
use free resources and reducing the cost of data acquisition.
Figure 2 shows the result after a WMS is opened in QGIS.
Figure 1 The major steps of digitizing process.
3.1 Preparation of referenced data source
Since the digitized results are location information extracted
from the referenced materials by human operators, the
specification, property and condition of the referenced materials
have dominant influence on the digitized coordinates. Students
must have thorough understanding about the technical
limitation and distinct properties of the referenced materials.
Four types of referenced material are used in this lab:
Scan of existing maps
To extract useful information from paper-based maps, a
commonly used approach is to scan the maps into digital
imagery and use heads-up digitizing approach to acquire the
location information. Students are first required to identify the
scale, time, context, coordinate reference system (CRS) of the
maps. The influence introduced in the scanning process, e.g.,
resolution, scanner calibration, etc., is also discussed. For those
map sheets with no CRS information, students are also required
to identify control points from the scanned images for the later
process of geo-registration.
Figure 2 Example of digitizing based WMS (QGIS).
Since the results of a digitizing process is stored in vector data
format. The teaching goal in this step is to help students to gain
a better understanding about the property of various available
resources, such that they have knowledge about the limitation of
referenced material and the procedure for selecting and
preparing for the best referenced material available.
Digital RS images
Digital RS images are widely used in current GIS-based
applications because of their advantages for offering up-to-date
information about reality. Most of the imagery products are
already geo-referenced to a specific CRS by their producers, so
the procedure of coordinate transformation and selection of
control points can be often skipped. Students, however, must be
aware of that the ground resolution, time and spectrum will
3.2 Digitizing procedure
The digitizing procedure in this lab is restricted to the
digitization performed by human operators. A step-by-step
87
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
guide for the digitization procedures is first introduced. Four
GIS software, namely, MapInfo, AutoDesk Map, Google Earth
and QGIS, are used in this lab. Because students are already
familiar with the operations of MapInfo in the course of “Basic
principles of GIS”, the other software is chosen to increase
students’ knowledge about their distinguished design of
digitizing functions. Based on their operating experience in the
lab, students are asked to give evaluation about the pros and
cons for the chosen software. The following questions must be
discussed in their reports:
1. Select the best GIS software to complete the task;
2. Design the most effective and efficient digitizing
procedures;
3. Reduce repetitive digitizing as much as possible;
4. Control the negative factors to the digitized results;
5. Ensure the quality of the digitized results.
3.3 Data quality
Data quality information serves as the basis for users to
determine if the data fits their application needs. As such
information can only be provided by people who really involve
in the process of producing data, every professional surveyor
must learn how to correctly evaluates the quality of the digitized
data and unambiguously convey data quality information to
users in a standardized way. The categorization of data quality
in ISO19113 (ISO, 2002) is introduced and three data quality
sub elements, completeness, positional accuracy and logical
consistency, are emphasized in this lab:
Q: How will the choice of control points affect the digitizing
results?
Q: How to detect the error in the registration of control points?
Q: How to setup the parameter of snap tolerance during
digitization?
Q: How to reduce duplicated digitizing loading?
Q: How to avoid the co-boundary problem?
Q: How to improve the efficiency of the digitizing task?
Q: How to ensure all the required data has been digitized?
Q: How to ensure the required topological constraints is met?
Q: How to deal with the transformation between different
coordinate reference systems?
Completeness (omission and commission)
Completeness denotes the evaluation if all the selected features
from the reference materials are digitized without loss (omission)
and if no features in the digitized data are included by mistakes
(commission). Students are required to evaluate the
completeness status of their digitized data by comparing it to
the reference dataset. Students in the same group are asked to
digitize the same reference material and compare their digitized
results. In this step, students learn how to develop a
standardized operating procedure for evaluate data quality
either by self-checking or reference data. Omission and
Commission are evaluated separately.
For example, the following two tasks are specifically designed
for students to understand the influence caused by the selection
of control points:
Task 1
Given a scanned image of a grid, students are required to
complete the following procedures:
1. Carefully digitize the corner points of the grid and
specify the correct coordinates of the corner points.
2. Digitize all the grid points and compare the digitized
results with their theoretical value.
3. Calculate the RMS based on the difference of the
comparisons in step 2.
Positional accuracy (absolute positional accuracy)
Absolute positional accuracy denotes the difference between the
digitized coordinate and the coordinate that is regarded as true
values. Students are asked to make comparisons between their
digitized coordinates and the coordinates of the check points
and select appropriate data quality measure from ISO19138
(ISO, 2006) to describe its data quality status.
This task uses a scanned image of grid to offer a theoretical
framework as the reference for digitization. Since every
digitized coordinate can be compared with its theoretical value,
it helps students to understand how accurate their digitizing
result can be.
Logical consistency (topological consistency)
Many applications require the digitized dataset to conform to
specific topological constraints, for example, polygons do not
overlap with each other, node-edge structure, etc. In addition to
the functions students learn to ensure the quality of the digitized
results in section 3.2, e.g., the handling of co-boundary
situation, students further learn how to verify the data quality
status by using specific functions from GIS software. For
example, the MapInfo provides a function to detect the overlaps
and gaps between polygons. Such tiny polygons can be
automatically found and eliminated from the digitized datasets.
To ensure the submitted results will meet the specified
topological constraints, students must master the skills of
selecting and using the variety of topological operations offered
by different GIS software.
Task 2
With the same scanned grid image, this task ask students to first
change the coordinate of one corner points by adding a
hypothesized error and then digitize the grid points one more
time. When compared with the results of task 1, the major error
now comes from the hypothesized error. Students are asked to
analyze if such an error can be found during digitization and
how to avoid such types of error.
Some of the above GIS software offers specifically designed
function to track the co-boundary of two neighbouring polygons,
such that they do not need to be digitized twice. Students are
required to test and evaluate how this type of function can help
to improve the efficiency of digitization. To gain more
knowledge about the variety of reference material, students are
required to first compare the specification of ortho-images,
Google Earth and RS images, then further compare the digitized
results from these three referenced materials and discuss the
major reasons for causing such differences. When handling a
digitizing task, we believe a professional surveyor must be able
to
To successfully convey the data quality information to users,
professional surveyors are required to be familiar with the way
quality information is created and distributed in geospatial
community. This step is designed to guide students to integrate
the knowledge they learned from courses of adjustment,
engineering statistics and geospatial data quality.
88
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
be added to improve students’ ability to meet the different
demands from geospatial applications.
3.4 Metadata.
The establishment of standardized metadata is a mandatory
requirement for every newly produced geospatial data, so it
should also be a mandatory skill to professional surveyors.
Compliant with the ISO19115 (ISO, 2003) standard, Taiwan
Spatial Metadata Profile (TWSMP) serves as the standard for
creating metadata. The metadata entry system (Figure 3) offered
by the TGOS (Taiwan Geospatial One Stop) is used for students
to establish standardized metadata (TGOS, 2011). In this lab,
we particularly focus on the establishment of metadata elements
related to data quality. Students are required to carefully
summarize the lineage information of digitized data step by step
via the metadata elements of process step and cited sources. As
a digitizing task will involve a series of procedures and at least
one reference material, this lab is a perfect example to let
student build a comprehensive understanding about the
metadata elements of data quality. Because the status of
completeness, absolute positional accuracy and topological
consistency has also been evaluated, students also learn how to
record their evaluated results with quantitative data quality
elements.
5. REFERENCE
ISO, 2002. ISO19113, Geographic Information - Quality
Principles.
ISO, 2003. ISO19115, Geographic Information - Metadata
ISO, 2006. ISO19138 – Geographic Information - Data Quality
Measure
Chang, K.-T., 2003. Introduction to GIS, 2nd edition, McGrawHill.
TGOS,
2011.
Taiwan
(http://tgos.nat.gov.tw/tgos/)
Geospatial
One
Stop
OGC, 2006. OpenGIS Web Map Server Implementation
Specification, v. 1.3.0.
Figure 3 Interface for metadata entry (TGOS).
Following the above procedure, students learn how to make a
plan for a digitizing assignment and convey information that is
useful to users with standardized metadata. In the course of
“GIS Lab”, all labs designed for producing or processing
geospatial datasets will follow a similar approach, such that
students can incrementally increase their knowledge and
experience.
4. CONCLUSION
To meet the variety of application demands from the geospatial
community, professional surveyors must be aware of the
changes to their role and responsibility in the future data
sharing and distribution environment. In addition to the solid
education and practices they receive from geodesy and remote
sensing related courses, we argued that the curriculum design
must extend to particularly include issues of data quality and
metadata and these two issues must be logically integrated with
the training of the production of geospatial data. We
demonstrate how we use free on-line resource and various GIS
software to help students building thorough knowledge about
all the aspects of digitization tasks. More courses dealing with
the collaboration of knowledge across different domains should
89
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
GIS AWARENESS FOR SURVEYORS –
A SPATIAL DATA DISTRIBUTION PERSECTIVE
Jung-Hong Hong a, *, Hone-Jay Chu a
a
Dept. of Geomatics, National Cheng-Kung University, Tainan, Taiwan – (junghong, honejay)@mail.ncku.edu.tw
KEY WORDS: surveyor, education, digitizing, geospatial data, metadata
ABSTRACT:
The rapid technological growth of GIS, internet and mobile devices has steered a variety of innovated and pioneering applications in
the past few years. All of these applications require spatial data that meet specified criteria to ensure the quality of the service content.
The traditional role for surveyors was to use surveying technology and instruments to determine the location of phenomena of
interests and offer map products to serve as the basis for decision making. As the application demands becomes more variety and
instantaneous, the role and responsibility for modern surveyors must be examined. Professional surveyors nowadays must have both
the knowledge and skills to produce geospatial data that precisely meet the various application needs and correctly convey their
understanding about data to domain users to avoid wrong decision making. This paper discussed how we include such considerations
into our curriculum design at the Dept of Geomatics, NCKU. During their studies, students are expected to establish a correct
understanding about the role and responsibility of surveyors and develop themselves to be “solution providers” for providing reliable
geospatial data.
1.
applications because it has dominant influence on the use of the
application. To produce geospatial data that precisely meet the
application needs is always regarded as the responsibility and
the first priority to surveyors. Surveyors are thus required to
receive all kinds of professional training, such that they have
sufficient professional knowledge to make a survey plan and
ensure the criteria about the quality of the products can be met.
A rather restricted viewpoint about the responsibility of
surveyors is surveyors turn in the final products (e.g., maps) and
leave the remaining works to the organizations that intend to
use the data. In the current data sharing environment, this
viewpoint is no longer valid because users nowadays can easily
search “all” of the available geospatial data, regardless of what
their original purpose is. Despite that datasets are still created
according to specific purposes, there is no way to predict how it
is going to be used. It becomes very risky if users cannot have a
correct understanding about the data they access and use. From
the perspective of data production, it is hence necessary to
examine how surveyors cope with the vast varieties and fast
growing of needs from the geospatial community. This paper
discusses our viewpoint regarding the role and responsibility
change of surveyors, and further uses the lab of the digitization
of geospatial data as an example to demonstrate how this
viewpoint is incorporated in our curriculum design. The
remaining of this paper is organized as follows: section 2 offers
a brief overview about the GIS education in the Dept. of
Geoamtics at NCKU and our viewpoints about the links
between surveyors and GIS-based applications. Section 3 uses
the topics of digitizing and distribution of geospatial data in the
GIS lab as an example to demonstrate our design approach.
Finally, section 4 summarized our major findings.
INTRODUCTION
The rapid technological growth of GIS, internet and mobile
devices has steered a variety of innovated and pioneering
applications in the past few years. For example, even an
inexperienced user can now easily access the electronic maps
and satellite images of an unfamiliar place via internet-based
service platforms like Google Earth or Bing Maps. As internet
resources provide “free” information to unrestricted domains of
users in such a convenient way, the education of geomatics,
regardless what levels they are, have forever changed. Another
interesting example is the overwhelming expansion of mobile
telecommunication markets, especially the success of those
technology based upon smart phones. By simply downloading
an APP, we can use smart phones to easily acquire location
information with built–in GPS chip and record other types of
information with other built-in devices like camera, voice
recorder or touchpad. For many applications, this is already
sufficient to serve as a mobile device for collecting geospatial
data, not to mention it can be also used in Location-Based
Service (LBS), where service content from remote servers are
retrieved based on the current location of users via internet
connection. These technology innovations surely give a
tremendous boost to the evolution of education in schools, as
the use of data, software and hardware all become much more
convenient and inexpensive than before. The most important
breakthrough is that this impact is not restricted to professional
people who dedicate to the theory, implementation and
applications of geospatial technology, any people, any
department, any organization who has the need of using
geospatial data can all be benefited by the new technology.
Geospatial data is a common and essential component that
needs to receive tremendous attention in the above innovated
* Corresponding author.
85
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
2. CORE OF CURRICULUM DESIGN
Year
sophomore
sophomore
junior
junior
junior
junior
2.1 Brief history of the Dept. of Geomatics at NCKU
The Dept. of Geomatics at NCKU was formerly the Dept. of
Surveying Engineering. Established in the year of 1978, Dept.
of Surveying Engineering was dedicated to the education of
professional surveyors. Students receive solid training from a
variety of surveying courses, e.g., plane surveying, geodesy,
photogrammetry, cartography, etc., to enrich their professional
knowledge and become professional surveyors. As the scope of
geospatial technology evolves with the progress of internet,
remote sensing and GIS, more advanced courses were added
into the curriculum framework. In 2004, the title of the
department was officially changed to the Dept. of Geomatics to
vividly reflect our major missions for dedicating to the
education and training of geospatial professionals. The current
domain of interests can be generally categorized into three
major domains: (1) geodesy and satellite positioning; (2)
photogrammetry and remote sensing and (3) geographic
information system. Students are required to take fundamental
courses from all three domains, and are encouraged to further
pursue advanced studies in any fields they choose. Every
domain has its own course map for students to follow. With a
strong and excellent foundation in surveying engineering
education, becoming a licensed surveyor is often regarded as a
minimal requirement for the alumni of this department. Students
who have strong commitments and research potential to pursue
advanced knowledge about geomatics can continue their
students in the master and Ph D. programs.
Course title
Introduction to cartography
Basic principles of GIS
Applications of GIS
Digital Mapping
GIS Lab
GIS data quality, standards and sharing
Table 1. GIS-related courses
According to the course maps of the department, students are
expected to complete all the required courses of the three major
domains after their junior years. Some students may continue to
pursue advanced studies in GIS fields under the supervision of
teachers. These students are trained to be able to apply for job,
e.g., project manager, engineer, etc., offered by private GIS
consulting firms. Other students may choose to pursue their
career as surveyors. Based on the training from GIS-related
courses, they are able to use GIS software as tools for the
handling and processing of geospatial datasets. Regardless what
their choice of professional career is, a correct understanding
about the production and distribution of geospatial data is the
key to our curriculum design.
2.3 Vision
Via the designed courses, every student graduated from this
department is expected to have sufficient professional
knowledge and ample hand-on experience to become a
professional licensed surveyor. From the perspective of GIS,
every student must have a clear and unambiguous
understanding about how geospatial data is produced and
distributed in the future data sharing environment. Our goals are
therefore designed as the following:
2.2 GIS curriculum
To suffice the needs of GIS training, various topics of GISrelated courses are offered in undergraduate studies (Table 1).
Undergraduate students are required to take at least 6 credits of
hour of courses from each domain to ensure they have broad
knowledge about how geospatial technology is developed and
evolved. Since the other two domains are mainly focusing on
the acquisition and processing of geospatial data, a link between
courses of these two domains and those of GIS must be
established and emphasized beforehand. The course
“Introduction to Cartography” serves as the foundation for all
applications that use maps as the illustration tools. By
introducing the basic principles of map making, students learn
how to design their own maps that can adapt to application
needs and produce maps following predefined mapping
specifications. The course of “Basic principles of GIS” offers
the basic understanding about GIS. Following the lifecycles of
GIS development, we respectively introduce the knowledge
about the acquisition, editing, processing, management, analysis,
distribution and application of geospatial data. Based upon this
basis, the course of “GIS applications” further expands
students’ views regarding how geospatial data can be integrated
and used in different applications. In addition to these three
required courses, the course of “Digital mapping” is designed to
address the use of computers on the production of maps in
digital format. The course of “GIS Lab” is designed to offer
more practical training on different aspects of GIS. The course
of “GIS data quality, standard and sharing” focuses on the
knowledge to facilitate the sharing of geospatial data via
standardized technology Students are expected to develop
themselves to be solution providers and team leaders that can
actively evaluate the best strategy to the given tasks.
1. Students must have the ability to analyze the
requirement of tasks at hands with professional knowledge,
2. Students must be able to offer the best strategy by
taking advantages of tools available to them, both
surveying and GIS.
3. Students must be able to evaluate the quality of their
products and the control of error factors during data
production.
4. Students must be able to convey their understanding
about data to users to avoid wrong decision making.
As an individual course may cover one or two aspects only, we
believe a more comprehensive design that encompass all the
above four goals in a logical and integrated fashion needs to be
introduced in our curriculum. Section 3 demonstrates how we
incorporate this requirement in our GIS lab design.
3. CURRICULUM DESIGN
Based on the visions raised in section 2.3, the following
discusses the lab design for “Digitizing of geospatial data”.
Digitizing was one of the fundamental GIS skills that allow
users to extract the spatial information of phenomena of
interests from existed medium, e.g., maps and imagery (Chang,
2003). The digitized location information is stored in digital
formats for further processing in GIS-based applications.
Typical examples in the earlier years were mainly paper maps,
e.g., historical maps, topographic maps, cadastral maps, etc.
After remote sensing images become powerful resources for
collecting instantaneous information of reality and reliable
reference for analyzing changes over time, the ability of
digitizing becomes a mandatory skill to GIS users. However,
86
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
largely restrict the outcomes of digitizing. As most of the
imagery is produced following specific technical specification,
commonly used imagery types in Taiwan are introduced.
most users simply learn how to “use” the digitizing function
without a correct understanding about the factors that may
influence the quality of the products. To educate surveyors to
offer high-quality digitized geospatial data for later applications,
the following factors are included in this lab:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Google Earth
Google Earth has received increasing attention from both
geospatial professionals and ordinary users becasue it not only
provides global coverage of the earth, but also historical
archives of images at the same place. Furthermore, the use of
Google Earth is free to the public. Google Earth thus quickly
becomes a popular and inexpensive source from which digitized
information can be collected. Despite its advantages, the
drawback for Google Earth is its inconsistent quality at different
places and users generally are not offered sufficient information
to determine its fitness for use. When users are not sure whether
the referenced material is reliable or not, the digitized result is
also questionable.
The basic workflow of digitizing process
The determination of coordinate reference system
The choose of common control points
The geo-registration procedures
The distinct property of various types of data sources
The efficient digitizing procedures
The evaluation of positional accuracy
The topological constraints of digitized data
The completeness constraints of digitized data
The use of free internet resources
The establishment of standardized metadata
The digitizing process in this lab is composed of four
major steps (Figure 1):
Web Map Service
Web Map Service (OGC, 2006) has quickly become an easy
way for organizations to publish their data to the internet. With
its imagery-based nature, users can easily acquire data from the
servers in imagery format and digitize phenomena of interests
with tools available from clients’ software. As WMS requires
providers to clearly specify its referenced CRS, the procedure is
very similar to the procedure for digitizing a RS image. Unlike
the above three types of sources, the image returned can be
arbitrarily specified by users and can be even superimposed in
Google Earth for further comparison. The National Center for
High-Performance Computing (NCHC) provides a variety of
WMS based on the RS images it collected. Unlike Google Earth,
the WMS from NCHS is dedicated to the images of Taiwan.
Some of them are the images collected after major natural
hazards. This serves as an outstanding resource for students to
use free resources and reducing the cost of data acquisition.
Figure 2 shows the result after a WMS is opened in QGIS.
Figure 1 The major steps of digitizing process.
3.1 Preparation of referenced data source
Since the digitized results are location information extracted
from the referenced materials by human operators, the
specification, property and condition of the referenced materials
have dominant influence on the digitized coordinates. Students
must have thorough understanding about the technical
limitation and distinct properties of the referenced materials.
Four types of referenced material are used in this lab:
Scan of existing maps
To extract useful information from paper-based maps, a
commonly used approach is to scan the maps into digital
imagery and use heads-up digitizing approach to acquire the
location information. Students are first required to identify the
scale, time, context, coordinate reference system (CRS) of the
maps. The influence introduced in the scanning process, e.g.,
resolution, scanner calibration, etc., is also discussed. For those
map sheets with no CRS information, students are also required
to identify control points from the scanned images for the later
process of geo-registration.
Figure 2 Example of digitizing based WMS (QGIS).
Since the results of a digitizing process is stored in vector data
format. The teaching goal in this step is to help students to gain
a better understanding about the property of various available
resources, such that they have knowledge about the limitation of
referenced material and the procedure for selecting and
preparing for the best referenced material available.
Digital RS images
Digital RS images are widely used in current GIS-based
applications because of their advantages for offering up-to-date
information about reality. Most of the imagery products are
already geo-referenced to a specific CRS by their producers, so
the procedure of coordinate transformation and selection of
control points can be often skipped. Students, however, must be
aware of that the ground resolution, time and spectrum will
3.2 Digitizing procedure
The digitizing procedure in this lab is restricted to the
digitization performed by human operators. A step-by-step
87
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
guide for the digitization procedures is first introduced. Four
GIS software, namely, MapInfo, AutoDesk Map, Google Earth
and QGIS, are used in this lab. Because students are already
familiar with the operations of MapInfo in the course of “Basic
principles of GIS”, the other software is chosen to increase
students’ knowledge about their distinguished design of
digitizing functions. Based on their operating experience in the
lab, students are asked to give evaluation about the pros and
cons for the chosen software. The following questions must be
discussed in their reports:
1. Select the best GIS software to complete the task;
2. Design the most effective and efficient digitizing
procedures;
3. Reduce repetitive digitizing as much as possible;
4. Control the negative factors to the digitized results;
5. Ensure the quality of the digitized results.
3.3 Data quality
Data quality information serves as the basis for users to
determine if the data fits their application needs. As such
information can only be provided by people who really involve
in the process of producing data, every professional surveyor
must learn how to correctly evaluates the quality of the digitized
data and unambiguously convey data quality information to
users in a standardized way. The categorization of data quality
in ISO19113 (ISO, 2002) is introduced and three data quality
sub elements, completeness, positional accuracy and logical
consistency, are emphasized in this lab:
Q: How will the choice of control points affect the digitizing
results?
Q: How to detect the error in the registration of control points?
Q: How to setup the parameter of snap tolerance during
digitization?
Q: How to reduce duplicated digitizing loading?
Q: How to avoid the co-boundary problem?
Q: How to improve the efficiency of the digitizing task?
Q: How to ensure all the required data has been digitized?
Q: How to ensure the required topological constraints is met?
Q: How to deal with the transformation between different
coordinate reference systems?
Completeness (omission and commission)
Completeness denotes the evaluation if all the selected features
from the reference materials are digitized without loss (omission)
and if no features in the digitized data are included by mistakes
(commission). Students are required to evaluate the
completeness status of their digitized data by comparing it to
the reference dataset. Students in the same group are asked to
digitize the same reference material and compare their digitized
results. In this step, students learn how to develop a
standardized operating procedure for evaluate data quality
either by self-checking or reference data. Omission and
Commission are evaluated separately.
For example, the following two tasks are specifically designed
for students to understand the influence caused by the selection
of control points:
Task 1
Given a scanned image of a grid, students are required to
complete the following procedures:
1. Carefully digitize the corner points of the grid and
specify the correct coordinates of the corner points.
2. Digitize all the grid points and compare the digitized
results with their theoretical value.
3. Calculate the RMS based on the difference of the
comparisons in step 2.
Positional accuracy (absolute positional accuracy)
Absolute positional accuracy denotes the difference between the
digitized coordinate and the coordinate that is regarded as true
values. Students are asked to make comparisons between their
digitized coordinates and the coordinates of the check points
and select appropriate data quality measure from ISO19138
(ISO, 2006) to describe its data quality status.
This task uses a scanned image of grid to offer a theoretical
framework as the reference for digitization. Since every
digitized coordinate can be compared with its theoretical value,
it helps students to understand how accurate their digitizing
result can be.
Logical consistency (topological consistency)
Many applications require the digitized dataset to conform to
specific topological constraints, for example, polygons do not
overlap with each other, node-edge structure, etc. In addition to
the functions students learn to ensure the quality of the digitized
results in section 3.2, e.g., the handling of co-boundary
situation, students further learn how to verify the data quality
status by using specific functions from GIS software. For
example, the MapInfo provides a function to detect the overlaps
and gaps between polygons. Such tiny polygons can be
automatically found and eliminated from the digitized datasets.
To ensure the submitted results will meet the specified
topological constraints, students must master the skills of
selecting and using the variety of topological operations offered
by different GIS software.
Task 2
With the same scanned grid image, this task ask students to first
change the coordinate of one corner points by adding a
hypothesized error and then digitize the grid points one more
time. When compared with the results of task 1, the major error
now comes from the hypothesized error. Students are asked to
analyze if such an error can be found during digitization and
how to avoid such types of error.
Some of the above GIS software offers specifically designed
function to track the co-boundary of two neighbouring polygons,
such that they do not need to be digitized twice. Students are
required to test and evaluate how this type of function can help
to improve the efficiency of digitization. To gain more
knowledge about the variety of reference material, students are
required to first compare the specification of ortho-images,
Google Earth and RS images, then further compare the digitized
results from these three referenced materials and discuss the
major reasons for causing such differences. When handling a
digitizing task, we believe a professional surveyor must be able
to
To successfully convey the data quality information to users,
professional surveyors are required to be familiar with the way
quality information is created and distributed in geospatial
community. This step is designed to guide students to integrate
the knowledge they learned from courses of adjustment,
engineering statistics and geospatial data quality.
88
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-6/W27, 2011
ISPRS Taipei 2011 Workshop, 4-6 October 2011, Taipei, Taiwan
be added to improve students’ ability to meet the different
demands from geospatial applications.
3.4 Metadata.
The establishment of standardized metadata is a mandatory
requirement for every newly produced geospatial data, so it
should also be a mandatory skill to professional surveyors.
Compliant with the ISO19115 (ISO, 2003) standard, Taiwan
Spatial Metadata Profile (TWSMP) serves as the standard for
creating metadata. The metadata entry system (Figure 3) offered
by the TGOS (Taiwan Geospatial One Stop) is used for students
to establish standardized metadata (TGOS, 2011). In this lab,
we particularly focus on the establishment of metadata elements
related to data quality. Students are required to carefully
summarize the lineage information of digitized data step by step
via the metadata elements of process step and cited sources. As
a digitizing task will involve a series of procedures and at least
one reference material, this lab is a perfect example to let
student build a comprehensive understanding about the
metadata elements of data quality. Because the status of
completeness, absolute positional accuracy and topological
consistency has also been evaluated, students also learn how to
record their evaluated results with quantitative data quality
elements.
5. REFERENCE
ISO, 2002. ISO19113, Geographic Information - Quality
Principles.
ISO, 2003. ISO19115, Geographic Information - Metadata
ISO, 2006. ISO19138 – Geographic Information - Data Quality
Measure
Chang, K.-T., 2003. Introduction to GIS, 2nd edition, McGrawHill.
TGOS,
2011.
Taiwan
(http://tgos.nat.gov.tw/tgos/)
Geospatial
One
Stop
OGC, 2006. OpenGIS Web Map Server Implementation
Specification, v. 1.3.0.
Figure 3 Interface for metadata entry (TGOS).
Following the above procedure, students learn how to make a
plan for a digitizing assignment and convey information that is
useful to users with standardized metadata. In the course of
“GIS Lab”, all labs designed for producing or processing
geospatial datasets will follow a similar approach, such that
students can incrementally increase their knowledge and
experience.
4. CONCLUSION
To meet the variety of application demands from the geospatial
community, professional surveyors must be aware of the
changes to their role and responsibility in the future data
sharing and distribution environment. In addition to the solid
education and practices they receive from geodesy and remote
sensing related courses, we argued that the curriculum design
must extend to particularly include issues of data quality and
metadata and these two issues must be logically integrated with
the training of the production of geospatial data. We
demonstrate how we use free on-line resource and various GIS
software to help students building thorough knowledge about
all the aspects of digitization tasks. More courses dealing with
the collaboration of knowledge across different domains should
89